The goal of the proposed study is to investigate the variables associated with use/non-use of subsidized respite care programs designed to assist caregivers of victims of Alzheimer's Disease and related disorders. Three sets of independent variables have been identified for investigation: (1) demographic characteristics, (2) factors associated with objective need; and (3) aspects of service delivery. A structural model of the relationships among the variables is proposed which posits reciprocal effects of respite use on critical variables such as mental stress experienced by the caregiver, functional abilities of the Alzheimer's victim, physical health of the caregiver, and types of health care services used. These variables are hypothesized to affect the caregiver's willingness or ability to continue providing care. A major goal of the study is to assess the adequacy of the hypothesized model. Funding for six model respite projects across the State of Michigan has been provided through the Michigan Department of Mental Health. Programs and service delivery vary according to the needs of participants in each area. A three-wave panel study is proposed which takes advantage of this existing program structure. Face-to-face interviews will be conducted with eligible program participants three times at four month intervals regardless of whether the participant has used respite services. Three types of caregivers will be represented in the sample: those who are using services ("users"), those who have inquired about services but never used them ("seekers"), and those caregivers who never seek out assistance ("non-seekers"). Information describing characteristics of the various programs, characteristics of the caregivers, and characteristics of the impaired elders will be collected. Linear structural equation and logistic regression approaches will be used to estimate structural parameters and assess the overall fit of the model. The results will be of direct use to program planners and health care professionals who assist families with Alzheimer's victims or other informal caregivers by: 1) providing information about when in the caregiving career respite is likely to be most effective; 2) showing how service delivery can be tailored to increase perceptions of utility of potential users; and 3) identifying caregivers who could likely benefit from respite but are unlikely to seek out such services on their own.